Refuse Collection Vehicle Route Management and Optimization

This application claims the benefit under 35 U.S.C. § 119(e) of U.S. Patent Application No. 63/679,417, entitled “REFUSE COLLECTION VEHICLE ROUTE MANAGEMENT AND OPTIMIZATION,” filed Aug. 5, 2024, which is incorporated herein by reference in its entirety.

This disclosure relates to systems and methods for route management and optimization for solid waste collection.

Refuse collection vehicles collect solid waste and transport the solid waste to landfills, recycling centers, or treatment facilities. Historically, refuse collection vehicles have employed diesel powered engines to propel the vehicle and a power takeoff (PTO) that provides hydraulic actuation for vehicle body systems. However, a developing demand for all-electric or partially-electric refuse vehicles has emerged. Improvements in the systems and methods for route management and optimization for solid waste collection, using all-electric or partially-electric refuse vehicles, are continually sought.

Implementations of the present disclosure are generally directed to refuse collection vehicles, systems, and methods to manage and optimize refuse collection vehicle routes using all-electric or partially-electric vehicles. More particularly, implementations of the present disclosure are directed to systems and methods configured to perform operations to determine a remaining capacity of a battery pack of a refuse collection vehicle and, based on the remaining capacity, restrict certain electric vehicle body functions when the remaining capacity reaches or falls below a threshold.

In an example implementation, a refuse collection vehicle includes: a chassis; a cab coupled to a front portion of the chassis; an electric vehicle body coupled to the chassis rearward of the cab, the electric vehicle body including one or more electrically powered body systems; at least one rechargeable battery pack configured to provide electric power to the body systems; and a computing device configured to perform operations including: calculating a state-of-charge of the at least one rechargeable battery pack based on a voltage of the at least one rechargeable battery pack, determining a remaining capacity of the at least one rechargeable battery pack, and restricting certain electric vehicle body functions when the remaining capacity falls below a threshold.

In some embodiments, the electric vehicle body includes one or more of an electrically-actuated tailgate, an electrically-actuated lift assembly, and an electrically-actuated refuse packing assembly.

In some embodiments, restricting certain electric vehicle body functions includes decreasing a speed of one or both of the electrically-actuated lift assembly and the electrically-actuated refuse packing assembly.

In some embodiments, decreasing the speed of the electrically-actuated lift assembly includes one or both of decreasing a speed at which an arm of the electrically-actuated lift assembly is lifted and decreasing a speed at which the arm is extended.

In some embodiments, the computing device is configured to generate a visual alert or an audible alert when the remaining capacity falls below the threshold.

In some embodiments, the threshold is determined adaptively.

In some embodiments, the computing device is an onboard computing device of the refuse collection vehicle.

In some embodiments, the refuse collection vehicle includes a display device within the cab, and the computing device is configured to display image data including one or both of the state-of-charge of the at least one rechargeable battery pack, and the remaining capacity of the at least one rechargeable battery pack.

In some embodiments, the computing device is configured to transmit the image data to one or more remote computing devices for display.

In some embodiments, the computing device is configured to perform operations further including: based on the remaining capacity, calculating a remaining number of refuse containers that can be collected on a refuse collection route.

In some embodiments, the computing device is configured to perform operations further including: comparing the remaining number of refuse containers that can be collected to a planned number of refuse containers to be collected, determining if the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected, and restricting certain electric vehicle body functions when the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected.

In some embodiments, the computing device is configured to generate a visual alert or an audible alert when the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected.

In some embodiments, the refuse collection vehicle includes a display device within the cab, and the computing device is configured to display image data including the remaining number of refuse containers that can be collected on a refuse collection route.

In some embodiments, the at least one rechargeable battery pack is mounted on the electric vehicle body.

In some embodiments, the at least one rechargeable battery pack is mounted on the chassis.

In some embodiments, the at least one rechargeable battery pack is a traction battery.

In an aspect combinable with the example implementation, a system includes: a refuse collection vehicle, including: a chassis; a cab coupled to a front portion of the chassis; an electric vehicle body coupled to the chassis rearward of the cab, the electric vehicle body including one or more electrically powered body systems; and at least one rechargeable battery pack configured to provide electric power to the body systems; and at least one processor communicably coupled to the electric vehicle body and the at least one rechargeable battery pack, the at least one processor configured to perform operations including: calculating a state-of-charge of the at least one rechargeable battery pack based on a voltage the at least one rechargeable battery pack, determining a remaining capacity of the at least one rechargeable battery pack, and restricting certain electric vehicle body functions when the remaining capacity falls below a threshold.

In another aspect combinable with any of the previous aspects, a computer-implemented method performed by at least one processor includes: calculating, by the at least one processor, a state-of-charge of at least one rechargeable battery pack of a refuse collection vehicle based on a voltage of the at least one rechargeable battery pack; determining, by the at least one processor, a remaining capacity of the at least one rechargeable battery pack; and restricting, by the at least one processor, certain functions of an electric vehicle body of the refuse collection vehicle when the remaining capacity falls below a threshold.

In some embodiments, the electric vehicle body includes one or more of an electrically-actuated tailgate, an electrically-actuated lift assembly, and an electrically-actuated refuse packing assembly.

In some embodiments, restricting certain electric vehicle body functions includes decreasing a speed of one or both of the electrically-actuated lift assembly and the electrically-actuated refuse packing assembly.

In some embodiments, decreasing the speed of the electrically-actuated lift assembly includes one or both of decreasing a speed at which an arm of the electrically-actuated lift assembly is lifted and decreasing a speed at which the arm is extended.

In some embodiments, the computer-implemented method further includes generating, by the at least one processor, a visual alert or an audible alert when the remaining capacity falls below the threshold.

In some embodiments, the threshold is determined adaptively.

In some embodiments, the computer-implemented method further includes displaying, by the at least one processor, on a display device within a cab of the refuse collection vehicle, image data including one or both of the state-of-charge of the at least one rechargeable battery pack and the remaining capacity of the at least one rechargeable battery pack.

In some embodiments, transmitting, by the at least one processor, the image data to one or more remote processors for display.

In some embodiments, the computer-implemented method further includes calculating, by the at least one processor, based on the remaining capacity, a remaining number of refuse containers that can be collected on a refuse collection route.

In some embodiments, the computer-implemented method further includes comparing, by the at least one processor, the remaining number of refuse containers that can be collected to a planned number of refuse containers to be collected, determining, by the at least one processor, if the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected, and restricting, by the at least one processor, certain electric vehicle body functions when the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected.

In some embodiments, the computer-implemented method further includes generating, by the at least one processor, a visual alert or an audible alert when the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected.

In some embodiments, the computer-implemented method further includes displaying, by the at least one processor, on a display device within a cab of the refuse collection vehicle, image data including the remaining number of refuse containers that can be collected on a refuse collection route.

Particular implementations of the subject matter described in this specification can be implemented so as to realize one or more of the following advantages.

For example, the systems, methods, and refuse collection vehicles of the present disclosure can advantageously determine a capacity of a battery pack of a refuse collection vehicle in real time. The refuse collection vehicles, systems, and methods of the disclosure can accomplish this by calculating a state-of-charge (SOC) of the battery pack based on continuous voltage measurements. Furthermore, the refuse collection vehicles, systems, and methods of the disclosure can optimize and manage the power of a refuse collection vehicle by using a processor or a computing device to determine a remaining capacity of the battery pack and to restrict certain functions of the refuse collection vehicle if the remaining capacity reaches or falls below a threshold, thereby conserving power.

In addition, the refuse collection vehicles, systems, and methods of the disclosure can alert the operator of the refuse collection vehicle if the remaining battery capacity is not sufficient to power the refuse collection vehicle through completion of a refuse collection vehicle route. This is achieved by using a processor or a computing device to calculate a remaining number of refuse containers to be collected on the refuse collection vehicle route and to compare the remaining number of refuse containers to be collected to a planned number of refuse containers to be collected. Moreover, the processor or the computing device can determine if the planned number of refuse containers to be collected exceeds the remaining number of refuse containers to be collected and consequently, can restrict vehicle body functions when the planned number of refuse containers to be collected exceeds the remaining number of refuse containers to be collected.

Thus, in some implementations, the systems, methods, and refuse collection vehicles of the present disclosure can advantageously manage and optimize battery usage to aid in the completion of a refuse vehicle collection route. Furthermore, in some implementations, the systems, methods, and refuse collection vehicles of the present disclosure can enable a customer to monitor the performance and efficiency of the refuse collection vehicle during a route. For example, a customer can track and compare the total time to complete a given route by different operators.

It is appreciated that methods in accordance with the present specification may include any combination of the aspects and features described herein. That is, methods in accordance with the present specification are not limited to the combinations of aspects and features specifically described herein, but also include any combination of the aspects and features provided.

The details of one or more implementations of the subject matter described in this disclosure are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the subject matter will be apparent from the description and drawings, and from the claims.

The refuse collection vehicles, systems, and methods of the present disclosure feature processors and/or computing devices to monitor and determine the capacity of a battery of a refuse collection vehicle in real time and perform certain actions (e.g., restrict vehicle body functions, determine and/or predict route completion, etc.) as a result. Furthermore, the processors and/or computing devices of the refuse collection vehicles described herein calculate the number of refuse containers that can be collected in a refuse collection route (i.e., the number of “picks”) based on the determined capacity of the battery in real time.

1 FIG. 100 100 102 136 102 102 102 106 106 108 102 102 illustrates an example systemfor the collection of refuse in accordance with the disclosed embodiments. The systemincludes a vehicleand one or more onboard computing devicescoupled to the vehicle. The vehiclecan be a refuse collection vehicle that operates to collect and transport refuse. The refuse collection vehicle can also be described as a garbage collection vehicle or garbage truck. The vehiclecan be configured to lift a containerthat contains refuse, and empty the refuse in the containerinto a hopperof the vehicle, to enable transport of the refuse to a collection site, compacting of the refuse, and/or other refuse handling activities. The vehiclecan also handle containers in other ways, such as by transporting the containers to another site for emptying.

102 110 102 113 113 110 112 114 112 116 112 114 118 120 122 108 118 120 122 122 108 114 108 116 102 124 126 128 126 128 116 1 FIG. 1 2 FIGS.and The vehiclecan include various body componentsthat are appropriate for the particular type of vehicle. For example, the vehicleis depicted as a front-loading refuse collection vehicle inand includes a vehicle bodythat is an all-electric vehicle body or a partially-electric vehicle body. The vehicle bodyincludes various body componentssuch as a chassis, a storage bodycoupled to a back portion of the chassis, and a cabcoupled to a front portion of the chassis. The storage bodyincludes a plurality of panels, a tailgate, and a hopper cover. The hopperis defined by the panels, the tailgate, and the hopper coverand includes a compartment that receives the collected refuse. The hopper coveris configured to cover the opening of the hopper. As shown in, the storage bodyand the hopperextend rearward of the cab. The vehicleincludes a lift assemblyfurther including a pair of opposing lift armsand a pair of opposing forks. The pair of lift armsand the pair of forksextend forward of the cab.

110 113 120 108 114 114 120 136 144 146 Some of the body componentscan be electrically-actuated. For example, the vehicle bodycan include an electrically-actuated tailgateto open and close (and optionally lock) the tailgate, an electrically-actuated refuse loading assembly (e.g., an electrically-actuated front-loading refuse collection vehicle, a side-loading refuse collection vehicle loader, or a rear-loading refuse collection vehicle), and an electrically-actuated refuse packing assembly configured transfer refuse from the hopperand to compact the refuse in the storage body. The packing assembly may be further configured to eject refuse from the storage bodywhen the tailgateis open. Other systems requiring electrical power may include, for example, vehicle and work lighting, the onboard computer deviceand camera(s)that are configured to generate image data.

113 115 115 120 115 113 112 102 102 113 112 102 113 112 102 102 115 115 113 115 113 117 102 115 113 115 112 The vehicle bodymay further include an electrical power source such as one or more battery packsthat may be rechargeable. In some embodiments, the battery packsdo not provide electrical power to at least one of the tailgate, a packer, or an auger. In some embodiments, the electrical power required by the packer and/or auger is averaged with a rolling average battery usage. In some embodiments, the battery packsinclude a first battery pack that provides electrical power to the vehicle bodyand a second battery pack that provides electrical power to the chassisof the vehicle. In some embodiments, the vehicleincludes a battery pack that provides electrical power to both the vehicle bodyand the chassisof the vehicle. In some embodiments, the operator can choose whether the vehicle bodyand the chassisof the vehicleshare a battery pack. For example, the option to share a battery pack is displayed on a screen of a display device of the vehicle. The battery pack(s)include one or more battery cells and may include substantially any suitable battery system, including but not limited to, nickel batteries, lithium batteries, and the like. The battery pack(s)may be deployed on any suitable location on the vehicle body. The battery pack(s)may be advantageously deployed on the underside of the vehicle body, for example, between adjacent frame rails, to lower the center of gravity of the vehicle. In other examples, battery pack(s)can be mounted on the vehicle body. In yet another example, the battery pack(s)can be mounted on the chassis.

113 115 115 102 115 The disclosed embodiments are not limited regarding the battery type or deployment in the vehicle body. For example, the battery pack(s)can be traction batteries. In some implementations, the battery pack(s)can be lithium ion batteries. Moreover, while not depicted, it will be understood that the vehiclemay further include an alternator electrically coupled with and configured to recharge the one or more battery pack(s).

112 112 113 The disclosed refuse vehicle embodiments may include vehicles including a natural gas-powered internal combustion engine and an all-electric body. In some implementations, the disclosed embodiments may include an all-electric vehicle including an electrically powered propulsion system (an electric motor) and electrically actuated body systems. In such embodiments the propulsion system and the body systems may receive power from the same electrical power source (e.g., located on the chassis) or from dedicated power sources (e.g., a first power source on the chassisand a second power source in the vehicle body). The disclosed embodiments are expressly not limited in this regard.

1 FIG. 102 124 108 102 It will be understood that the disclosed embodiments are not limited to any particular type or style of refuse vehicle. The vehicle may include a sanitation truck, a recycling truck, a garbage truck, a waste collection truck, etc. In, the depicted vehicleis configured as a front-loading refuse collection vehicle; including an electrically-actuated lift assemblyconfigured to load refuse into the hopperfrom alongside the vehicle. The disclosed embodiments are, of course, not limited in regard to any refuse loading configuration.

1 FIG. 1 FIG. 102 114 110 102 102 110 110 110 For example, while not depicted on, the vehiclemay be an all-electric or partially-electric side-loading refuse collection vehicle with an automated side loader (ASL) (e.g., a lift assembly that extends from a side of the storage body). A vehicle with an ASL may include body componentsinvolved in the operation of the ASL, such as an arm and/or grabbers as well as other body components, such as a pump, a tailgate, a packer, and so forth. In some examples, the vehiclemay be a side-loading refuse collection vehicle without a lift assembly. In some embodiments, the vehiclemay be an all-electric or partially electric rear-loading refuse collection vehicle that may include body components, such as a pump, blade, tipper, and so forth. A front-loading refuse collection vehicle, such as the example shown in, may include body components, such as a pump, tailgate, packer, fork assembly, commercial grabbers, and so forth. Body componentsmay also include other types of components that operate to bring garbage into a hopper of a refuse collection vehicle, compress and/or arrange the garbage in the refuse collection vehicle, and/or expel the garbage from the refuse collection vehicle.

102 124 128 106 106 126 128 113 1 FIG. It will further be understood that the particular electrically actuated body systems employed by a refuse collection vehicle may depend on the type and configuration of the refuse collection vehicle. For example, the vehicledepicted onincludes a front-loading lift assemblyand may include an electrically-actuated pair of forks(configured to be received in fork pockets of the containerto enable the lifting of the container) and an electrically-actuated pair of lift arms(configured to move the pair of forksinwardly and outwardly with respect to the vehicle bodyand up and down with respect to a ground surface).

102 130 110 132 110 130 110 110 110 130 132 130 114 102 110 130 The vehiclecan include any number of body sensor devicesthat sense body component(s)and generate body sensor datadescribing the operation(s) and/or the operational state of various body components. The body sensor devicesmay be arranged in the body components, or in proximity to the body components, to monitor the operations of the body components. The body sensor devicesemit signals that include the body sensor datadescribing the body component operations, and the signals may vary appropriately based on the particular body component being monitored. The body sensor devicescan be provided on the storage bodyof the vehicleto evaluate cycles and/or other parameters of various body components. For example, as described in further detail herein, the body sensor devicescan detect and/or measure the particular position and/or operational state of body components such a lift arm, a fork assembly, and so forth.

130 130 110 130 110 The body sensor devicescan include, but are not limited to, an analog sensor, a digital sensor, a Controller Area Network (CAN) bus sensor, a magnetostrictive sensor, a radio detection and ranging (RADAR) sensor, a light detection and ranging (LIDAR) sensor, a laser sensor, an ultrasonic sensor, an infrared (IR) sensor, a stereo camera sensor, a three-dimensional (3D) camera, in-cylinder sensors, or a combination thereof. In some implementations, the body sensor devicesmay be incorporated into the various body components. Alternatively, the body sensor devicesmay be separate from the body components.

130 110 102 130 126 128 130 126 128 106 108 102 130 102 130 126 130 128 130 126 130 One or more body sensor devicescan be situated to determine the state and/or detect the operations of the body components. In some embodiments, the vehicleincludes one or more body sensor devicesthat are arranged to detect the position of the pair of lift armsand/or the pair of forks. For example, the body sensor device(s)can provide data about the current position of the pair of lift armsand/or the pair of forksthroughout a cycle to dump refuse from the containerinto the hopperof the vehicle. In some implementations, the body sensor device(s)are located in one or more cylinders of the vehicle. In some examples, a first body sensor deviceis located inside a cylinder used for raising the pair of lift armsand a second body sensor deviceis located inside a cylinder used for moving the pair of forks. In some implementations, a body sensor deviceis located on the outside of a housing containing the cylinder coupled to a lift arm. In some examples, a body sensor deviceis an in-cylinder, magnetostrictive sensor.

132 130 136 102 136 136 102 132 130 136 130 136 130 136 130 136 132 132 136 In some implementations, the body sensor datamay be communicated from the body sensor devicesto the onboard computing devicein the vehicle. In some instances, the onboard computing deviceis an under-dash device (UDU), and may also be referred to as the “gateway.” Alternatively, the onboard computing devicemay be placed in some other suitable location in or on the vehicle. The body sensor datamay be communicated from the body sensor devicesto the onboard computing deviceover a wired connection (e.g., an internal bus) and/or over a wireless connection. In some implementations, a Society of Automotive Engineers J1939 standard bus, in conformance with International Organization of Standardization (ISO) standard 11898, connects the various the body sensor deviceswith the onboard computing device. In some implementations, a Controller Area Network (CAN) bus connects the various the body sensor deviceswith the onboard computing device. For example, a CAN bus in conformance with ISO standard 11898 can connect the body sensor deviceswith the onboard computing device. In some implementations, the body sensor datadigitize the signals that communicate the body sensor databefore sending the signals to the onboard computing deviceif the signals are not already in a digital format.

132 136 138 136 132 110 132 110 The analysis of the body sensor datacan be performed at least partly by the onboard computing device, e.g., by processes that execute on the processor(s). For example, the onboard computing devicecan execute processes that perform an analysis of the body sensor datato determine the current position of the body components, such as the lift arm position or the fork assembly position. In some implementations, an onboard program logic controller or an onboard mobile controller perform analysis of the body sensor datato determine the current position of the body components.

136 138 140 142 136 The onboard computing devicecan include one or more processorsthat provide computing capacity, data storageof any suitable size and format, and one or more network interface controller(s) (NIC(s))that facilitate communication of the onboard computing devicewith other device(s) over one or more wired or wireless networks.

102 102 102 114 102 102 136 In some implementations, the vehicleincludes a body controller that manages and/or monitors various body components of the vehicle. The body controller of the vehiclecan be connected to multiple sensors in the storage bodyof the vehicle. The body controller can transmit one or more signals over the J1939 network, or other wiring on the vehicle, when the body controller senses a state change from any of the sensors. These signals from the body controller can be received by the onboard computing devicethat is monitoring the J1939 network.

136 136 102 136 136 In some implementations, the onboard computing deviceis a multi-purpose hardware platform. The onboard computing devicecan include a under dash unit (UDU) and/or a window unit (WU) (e.g., camera) to record video and/or audio operational activities of the vehicle. The onboard computing device hardware subcomponents can include, but are not limited to, one or more of the following: a central processing unit (CPU), a memory or data storage unit, a CAN interface, a CAN chipset, NIC(s) such as an Ethernet port, USB port, serial port, I2c lines(s), and so forth, I/O ports, a wireless chipset, a global positioning system (GPS) chipset, a real-time clock, a micro SD card, an audio-video encoder and decoder chipset, and/or external wiring for CAN and for I/O. The onboard computing devicecan also include temperature sensors, battery and ignition voltage sensors, motion sensors, CAN bus sensors, an accelerometer, a gyroscope, an altimeter, a GPS chipset with or without dead reckoning, and/or a digital CAN interface (DCI). The DCI cam hardware subcomponent can include the following: a CPU, memory, CAN interface, CAN chipset, Ethernet port, USB port, serial port, I2c lines, I/O ports, a wireless chipset, a GPS chipset, a real-time clock, and external wiring for CAN and/or for I/O. In some implementations, the onboard computing deviceis a smartphone, tablet computer, and/or other portable computing device that includes components for recording video and/or audio data, processing capacity, transceiver(s) for network communications, and/or sensors for collecting environmental data, telematics data, and so forth.

2 FIG. 200 136 115 115 202 115 136 115 102 depicts a flow chart of an example methodby which the onboard computing devicecan calculate a rolling average of the remaining capacity of the battery pack(s)during a refuse vehicle collection route. The voltage in each battery cell of the battery pack(s)is averaged () to yield an average voltage per battery cell. In some embodiments, the voltage in each battery cell of the battery pack(s)is measured by a manufacturer of the batteries, and the measurements are transmitted to the onboard computer device. In alternative embodiments, the voltage in each battery cell of the battery pack(s)is measured by a manufacturer, an operator, or a device onboard the vehicle.

136 115 204 136 202 The onboard computing devicecalculates an initial state-of-charge (SOC) of the battery pack(s)(). In some embodiments, the onboard computing devicecalculates the initial SOC by calculating an actual battery voltage span, which is determined by subtracting a minimum battery cell voltage from the average voltage per battery cell determined in step. In some embodiments, the minimum battery cell voltage is a predetermined value.

136 136 102 102 206 102 Next, the onboard computing devicecalculates the initial SOC by dividing the actual battery voltage span by the number of volts per 1% SOC. The number of volts per 1% SOC is determined by dividing a total battery voltage span (e.g., the battery's minimum cell voltage subtracted from the battery's maximum cell voltage) by 100. The total battery voltage span is the total voltage span of a battery cell that is fully charged (e.g., charged to 100% of its capacity). In some embodiments, the number of volts per 1% SOC is a predetermined value. The onboard computing devicethen records the initial SOC before the vehicleperforms any container picks (e.g., before the vehiclecollects any refuse containers on a refuse vehicle collection route) (). As used herein, the term “pick” or “picks” refers to a refuse container or refuse containers, respectively, being collected by the vehicleon a refuse collection route.

136 208 136 136 136 All SOC, voltage, or other battery capacity-related calculations that were previously stored (e.g., from a previous day) in the onboard computing deviceare then zeroed () by the onboard computing device. In some embodiments, the SOC, voltage, or other battery capacity-related calculations that were previously stored (e.g., from a previous day) in the onboard computing deviceare zeroed automatically by the onboard computing device.

136 115 210 The onboard computing devicestores the SOC of the battery pack(s)() prior to collecting a refuse container.

136 212 214 Once the SOC value is stored, the onboard computing deviceproceeds to start () and complete () the collection of the refuse container.

136 115 216 The onboard computing devicethen calculates the SOC of the battery pack(s)after the collection of the refuse container has been completed and stores () this SOC value.

136 102 218 Next, the onboard computing devicebegins to count the number of refuse containers that have been collected by the vehicleand increments its counter by one after each refuse container is collected ().

136 220 115 115 115 136 115 115 The total battery usage after each refuse container is collected is calculated by the onboard computing device() by comparing the current SOC of the battery pack(s)to the maximum capacity of the battery pack(s)(e.g., when the battery pack(s)are fully charged or at the start or refuse collection route). For example, the onboard computing devicesubtracts the current SOC of the battery pack(s)from the maximum capacity of the battery pack(s).

136 222 102 The onboard computing devicecalculates an estimated battery usage per “pick” () by dividing the total battery usage up until that point by the total number of refuse containers that have been collected by the vehicleup until that point. In this example, the estimated battery usage per “pick” is a rolling average battery usage.

136 224 115 115 The onboard computing devicecalculates the battery usage for the last refuse container that was collected () by subtracting the SOC of the battery pack(s)at the end of the collection of the refuse container, at that point in time, from the SOC of the battery pack(s)at the beginning of the collection of that same refuse container.

136 102 226 The onboard computing devicethen requests an input from an operator of the vehicleprompting the question of the refuse collection route has been completed ().

102 136 228 136 102 136 210 115 If the operator of the vehicleprovides “yes” as an input, the onboard computing deviceproceeds to store the data (e.g., the average SOC usage and total SOC usage) generated during that day (). In addition, the onboard computing deviceproceeds to store the number of “picks” reflected by the counter. In some embodiments, the stored data can be accessed at a later time (e.g., at the start of a new refuse collection route the next day). If the operator of the vehicleprovides “no” as an input, the onboard computing deviceproceeds to stepto store the SOC of the battery pack(s)at the start of the collection of a new refuse container.

3 FIG. 300 136 136 115 302 115 102 136 115 102 115 206 200 depicts a flow chart of an example methodby which the onboard computing devicecan determine a remaining number of refuse containers to be collected in a refuse collection route. The onboard computing devicebegins by determining an initial SOC of the battery pack(s)(). In some embodiments, the initial SOC of the battery pack(s)is the SOC of the battery pack(s) before the vehiclecollects the first refuse container in the refuse collection route. In some embodiments, the onboard computing devicedetermines an initial SOC of the battery pack(s)by retrieving the initial SOC value from a memory of the computing system of the vehicle, as the SOC of the battery pack(s)is recorded and stored before collecting any refuse containers as described by stepof method.

136 115 304 136 115 202 204 200 Next, the onboard computing devicedetermines the current SOC of the battery pack(s)(). In some embodiments, the onboard computing devicedetermines the current SOC of the battery pack(s)by performing a calculation relating a voltage of the battery pack(s) to the SOC as described in stepsandof method.

136 115 115 306 115 102 115 304 The onboard computing devicethen determines total battery usage by subtracting the current SOC of the battery pack(s)from the initial SOC of the battery pack(s)(). In some embodiments, the total battery usage reflects the total charge of the battery pack(s)that has been used since the start of the refuse collection route (e.g., before the vehiclecollects the first refuse container in the refuse collection route) up until the point in time when the current SOC of the battery pack(s)is determined in step.

136 102 308 136 102 218 102 Next, the onboard computing devicedetermines the number of “picks” or the number of refuse containers that have been collected by the vehicle(). In some embodiments, the onboard computing devicedetermines the number of “picks” by retrieving this number from the memory of the computing system of the vehicle. As stepdescribes, a container pick counter counts the number of refuse containers that are collected, records this value, and stores it in the memory of the computing system of the vehicle.

136 310 136 306 102 308 136 136 The onboard computing devicethen determines a rolling average battery usage (). In some embodiments, the onboard computing devicedetermines a rolling average battery usage by dividing the total battery usage, determined in step, by the number of refuse containers that have been collected by the vehiclethat was determined in step. In some embodiments, the onboard computing devicedetermines the rolling average battery usage by continuously updating the calculation of the battery usage average to include all the data (e.g., total battery usage and number of “picks”) until a determined time point (e.g., until the most recent time point). In some embodiments, the onboard computing devicedetermines a rolling average battery usage in real time.

136 312 136 115 304 310 Next, the onboard computing devicedetermines an estimated remaining number of refuse containers to be collected based on the rolling average battery usage (). In some embodiments, the onboard computing devicedetermines the estimated remaining number of refuse containers to be collected by dividing the current SOC of the battery pack(s), determined in step, by the rolling average battery usage (e.g., having units such as % SOC per “pick”) that is determined in step.

136 115 In some embodiments, the onboard computing devicefurther adds a “buffer zone” to the estimated remaining number of refuse containers to be collected. In some embodiments, adding a “buffer zone” is done to provide increased safety and to account for potential variability in the current SOC of the battery pack(s)and the rolling average battery usage.

136 In some embodiments, the “buffer zone” is a predetermined number or a percentage of the number of refuse containers to be collected that can be added to and/or subtracted from the estimated remaining number of refuse containers to be collected. For example, in some embodiments, the onboard computing devicecan add and/or subtract a “buffer zone” of 20% (of the estimated remaining number of refuse containers to be collected) to the estimated remaining number of refuse containers to be collected. In some embodiments, the “buffer zone” can range from about 5% to about 30% (e.g., about 5% to about 10%, 5% to about 15%, 5% to about 20%, 5% to about 25%, or 15% to about 30%) of estimated remaining number of refuse containers to be collected.

136 136 136 136 Alternatively, in some embodiments, the onboard computing devicecan determine the “buffer zone.” For example, the onboard computing devicecan calculate an average and standard deviation of the estimated remaining number of refuse containers to be collected. Then, the onboard computing devicecan use the standard deviation as the “buffer zone” to be added to and/or subtracted from the estimated remaining number of refuse containers to be collected. In some embodiments, the onboard computing deviceincludes more than one standard deviation (e.g., two or three standard deviations) in the “buffer zone.”

136 102 115 In some embodiments, the onboard computing deviceprovides a rolling estimated remaining number of refuse containers to be collected that updates continuously after every “pick.” In some embodiments, the estimated remaining number of refuse containers to be collected provides the operator with an estimate of the number of refuse containers that can be collected by the vehiclewith the remaining charge of the battery pack(s)at a determined time point (e.g., at the time point when the current SOC of the battery pack(s) is determined). In some embodiments, the estimated remaining number of refuse containers to be collected is based on the average of refuse containers that have been collected from the start of the refuse collection route until a determined time point (e.g., until the time point when the current SOC of the battery pack(s) is determined).

4 FIG. 400 136 115 402 404 406 408 410 412 302 304 306 308 310 312 412 136 102 414 136 416 402 416 depicts a flow chart of an example methodby which the onboard computing devicecan determine a remaining capacity of the battery pack(s)and can further restrict certain electric vehicle body functions when the remaining capacity is determined to be insufficient to complete the refuse collection route, thereby optimizing the refuse collection route. Steps,,,,, andare substantially the same as steps,,,,, anddiscussed above. Based on the estimated number of remaining picks, which is determined in step, the onboard computing devicecompares the estimated remaining number of picks to the total number of picks that are planned to be collected on the refuse collection route of the vehicle(). Based on this comparison, the onboard computing devicedetermines whether the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected (). In some embodiments, steps-are performed in real-time (e.g., during the refuse collection route).

136 102 418 If it is determined that the planned number of refuse containers to be collected does not exceed the estimated remaining number of refuse containers that can be collected based on the remaining battery capacity, then the onboard computing devicedoes not initiate a change to a functionality of the vehicle().

136 102 If it is determined that the planned number of refuse containers to be collected exceeds the estimated remaining number of refuse containers that can be collected, then the onboard computing deviceperforms operations to restrict certain electric vehicle body functions of the vehicle.

136 124 420 For example, the onboard computing devicecan perform operations to restrict a speed of the electrically-actuated lift assembly().

136 422 In some examples, the onboard computing devicecan perform operations to restrict a speed of the electrically-actuated refuse packing assembly ().

136 124 136 124 136 In some embodiments, the onboard computing devicecan perform operations to restrict the speed of both electrically-actuated lift and refuse packing assemblies. In some embodiments, decreasing the speed of the electrically-actuated lift assembly includes decreasing a speed at which an arm of the electrically-actuated lift assemblyis lifted and/or decreasing a speed at which the arm is extended. In some embodiments, the onboard computing deviceperforms operations to shift the arm of the electrically-actuated lift assemblyinto a low-power mode to conserve power. In some examples, the onboard computing devicecan perform operations to restrict an actuation of the electrically-actuated refuse packing assembly.

136 400 The onboard computing devicemay use a method that is substantially similar in function in several aspects to the example methoddiscussed above, but can include an alternative method to restrict certain electric vehicle body functions. For example, in some embodiments, the alternative method can include restricting certain electric vehicle body functions when the remaining capacity falls below a threshold, instead of when the remaining capacity is determined to be insufficient to collect the planned number of refuse containers. In some embodiments, the threshold is determined adaptively. In some embodiments, the threshold can be about 10% battery capacity, such that the alternative method can include restricting certain electric vehicle body functions when the remaining capacity falls below about 10%. In some embodiments, the threshold can be about 20% battery capacity, such that the alternative method can include restricting certain electric vehicle body functions when the remaining capacity falls below about 20%. In some implementations, the threshold can be about 10% to about 20% battery capacity.

102 136 Among other things, the techniques described herein include a method for detecting outliers in the amount of battery usage per refuse container that is collected by the vehicle. In some embodiments, the method can use a machine learning (ML) model. ML models or techniques are also referred to herein as artificial intelligence (AI). The machine learning techniques described herein can be implemented on at least one computing processor (e.g., the onboard computing deviceor a cloud computing server) and/or at least one hardware accelerator coupled to the at least one computing processor.

102 115 For example, in some embodiments, the method includes the following steps. The method includes receiving a plurality of data points obtained from the vehicleduring the refuse collection route. In some embodiments, the plurality of data points include the amount of charge of the battery pack(s)that is used to collect each refuse container during the refuse collection route. In some embodiments, the plurality of data points have units such as % SOC per “pick.” In some embodiments, the amount of charge is determined by retrieving the rolling average battery usage at the time of the collection of each refuse container during the refuse collection route. In some embodiments, the plurality of data points include the amount of power used to collect each refuse container during the refuse collection route.

115 115 115 115 115 Next, the method includes comparing the amount of charge of the battery pack(s)that is used to collect each refuse container to one or more thresholds (e.g., a maximum threshold and a minimum threshold). In some embodiments, the maximum threshold is about at least 10% higher than the average amount of charge of the battery pack(s)that is used to collect each refuse container. In some embodiments, the minimum threshold is about at least 10% lower than the average amount of charge of the battery pack(s)that is used to collect each refuse container. In some embodiments, the average amount of charge of the battery pack(s)that is used to collect each refuse container is determined by averaging the plurality of data points of one refuse collection route. In some embodiments, the average amount of charge of the battery pack(s)that is used to collect each refuse container is determined by averaging the plurality of data points of two or more refuse collection routes (e.g., the refuse collection routes performed in one week, one month, two to six months, or six months to a year). In some embodiments, the maximum and minimum thresholds are determined by calculating a standard deviation of the plurality of data points. For example, in some embodiments, the maximum and minimum thresholds can be three standard deviations above and below the average, respectively. In some embodiments, the ML model uses a z-score to determine the maximum and minimum thresholds. In some embodiments, the ML model uses an interquartile range to determine the maximum and minimum thresholds. In some embodiments, the ML model uses percentiles to determine the maximum and minimum thresholds. For example, in some embodiments, the ML model can use a custom range that accommodates all data points that lie anywhere between a minimum and a maximum percentile of the dataset and can filter the data points using the minimum and maximum limits defined by the custom range.

136 Next, the method includes identifying each data point as either an outlier or not an outlier based on the comparison to the one or more thresholds. For example, the data point is identified as an outlier if it exceeds the maximum threshold or if it is under the minimum threshold. Such classification is also referred to as binary classification. In one implementation, the identification of each data point can be performed manually. In alternate implementations, the identification of each data point can be performed by the computing processor (e.g., the onboard computing deviceor a cloud computing server) in an automated fashion. For example, the computing processor can implement a statistical algorithm to calculate the average of the plurality of data points and determine the maximum and minimum thresholds (e.g., by calculating a standard deviation of the plurality of data points or using any of the aforementioned methods). The algorithm can then identify and classify an outlier if it exceeds the maximum threshold or if it is under the minimum threshold.

Next, the method includes training the ML model, based on the previous identifying step, using the plurality of data points. As the received data points have been classified, the actual output of the sample inputs (i.e., the received data points) are known. The machine learning model may, however, generate a different output. The difference between the known, correct output for the sample inputs and the actual output of the machine learning model is referred to as a training error. The purpose of the training of the ML model is to reduce the training error until the model produces an accurate prediction for the training set. Training is the process of learning (i.e., determining) weights and bias values that the ML model should apply when inferences are made while minimizing the error (i.e., inaccuracy) in making predictions. In some implementations, errors are minimized and biases are reduced by performing tests and comparisons between the ML model and the prototype.

102 102 Next, the method includes optimizing the ML model by performing learning against a validation set (e.g., a test set). The training set is a group of sample inputs to be fed into the ML model (e.g., a neural network model) to train the ML model, and the validation set is a group of inputs and corresponding outputs that are used to determine the accuracy of the ML model when the ML model is being trained. While the plurality of data points are described as being received from the vehicle, in other implementations the plurality of data points can be received from a two or more vehicles.

The computing processor can fine-tune (e.g., improve the accuracy of the parameters of the trained machine learning model by performing learning using the validation set. Such fine-tuning can also be referred to as optimization of the machine learning model. The fine-tuning (or optimization) can implement various optimization algorithms, such as gradient descent, stochastic gradient descent, mini-batch gradient descent, momentum, adaptive moment estimation (also referred to as Adam), and/or the like. The computing processor can implement an algorithm based on the computational aspects (e.g., computational architecture and structure) of the computing processor. The gradient descent algorithm advantageously involves simple computations and is easy to implement and easy to understand. The stochastic gradient descent algorithm advantageously involves frequent updates of model parameters and thus converges in less time, and requires less memory as there is no need to store values of loss functions. The mini-batch gradient descent algorithm advantageously frequently updates the model parameters, has less variance, and requires a medium amount of memory. The momentum algorithm advantageously reduces the oscillations and high variance of the parameters, and converges faster than gradient descent. The Adam algorithm advantageously is fast and converges rapidly, rectifies vanishing learning rate, and has a high variance.

Thus, in the training phase, a known data set is put through an untrained machine learning model (e.g., untrained neural network), the results are compared with known results of the data set, and the framework reevaluates the error value and updates the weight of the data set in the layers of the neural network based on accuracy of the value. This reevaluation advantageously adjusts the neural network to improve the performance of the specific task—i.e., the classification task of classifying a data point as being an outlier or not an outlier—that the neural network is learning.

115 115 The optimized ML model is then used to generate a prediction for an estimated remaining number of refuse containers to be collected based on the available battery SOC and based on the plurality of data points that have been classified and filtered (e.g., by excluding the outliers) by the ML model. The prediction thus prevents the outliers from skewing the estimated remaining number of refuse containers to be collected. For example, in some embodiments, the prediction prevents a heavier load that exceeds the maximum threshold and is collected to skew the estimated remaining number of “picks” based on the available SOC of the battery pack(s). This advantageously provides the operator with a more accurate estimate to complete the refuse collection route while maximizing energy usage of the battery pack(s). Unlike the training phase, the deployment phase does not reevaluate or adjust the layers of the neural network based on the results, and instead the prediction applies knowledge from the trained neural network model and a uses that model to predict the estimated remaining number of refuse containers to be collected. Therefore, when a new set of one or more data points is input through the trained neural network, the neural network model outputs a prediction of estimated remaining number of refuse containers to be collected based on the predictive accuracy of the neural network.

1 FIG. 102 148 116 102 148 150 136 102 150 148 136 115 136 136 Referring back to, the vehicleincludes a display deviceinside the cabof the vehicle. The display deviceincludes a screen. In some implementations, in response to determining that the remaining battery capacity falls below the threshold or that the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected, the onboard computing devicecan cause a visual alert and/or an audible alert to be generated that alerts an operator of the vehicleto the status of the remaining battery capacity. In some implementations, a visual alert is displayed on the screenof the display devicein response to determining that the remaining battery capacity falls below the threshold or that the planned number of refuse containers to be collected exceeds the remaining number of refuse containers that can be collected. In some embodiments, the onboard computing deviceis configured to display image data including the SOC of the battery pack(s)and/or the remaining battery capacity. In some embodiments, the onboard computing deviceis configured to display image data including the remaining number of refuse containers that can be collected on a refuse collection route. In some examples, the onboard computing deviceis configured to transmit the image data to one or more remote computing devices for display.

5 FIG. 500 500 136 500 depicts an example computing system, according to implementations of the present disclosure. The systemmay be used for any of the operations described with respect to the various implementations discussed herein. For example, the systemmay be included, at least in part, in one or more of the onboard computing device, and/or other computing device(s) or system(s) described herein. The systemis intended to include various forms of digital computers, such as printed circuit boards (PCB), processors, digital circuitry, or otherwise. Additionally, the system can include portable storage media, such as, Universal Serial Bus (USB) flash drives. For example, the USB flash drives may store operating systems and other applications. The USB flash drives can include input/output components, such as a wireless transmitter or USB connector that may be inserted into a USB port of another computing device.

500 510 520 530 540 510 520 530 540 510 500 510 The systemincludes a processor, a memory, a storage device, and an input/output device. Each of the components,,, andare interconnected using a system bus. The processoris capable of processing instructions for execution within the system. The processor may be designed using any of a number of architectures. For example, the processormay be a CISC (Complex Instruction Set Computers) processor, a RISC (Reduced Instruction Set Computer) processor, or a MISC (Minimal Instruction Set Computer) processor.

510 510 510 520 530 540 In one implementation, the processoris a single-threaded processor. In another implementation, the processoris a multi-threaded processor. The processoris capable of processing instructions stored in the memoryor on the storage deviceto display graphical information for a user interface on the input/output device.

520 500 520 520 520 The memorystores information within the system. In one implementation, the memoryis a computer-readable medium. In one implementation, the memoryis a volatile memory unit. In another implementation, the memoryis a non-volatile memory unit.

530 500 530 530 The storage deviceis capable of providing mass storage for the system. In one implementation, the storage deviceis a computer-readable medium. In various different implementations, the storage devicemay be a floppy disk device, a hard disk device, an optical disk device, or a tape device.

540 500 540 540 540 540 540 116 102 540 The input/output deviceprovides input/output operations for the system. In one implementation, the input/output deviceincludes a joystick. In some implementations, the input/output deviceincludes a display unit for displaying graphical user interfaces. For example in some implementations, the input/output deviceis a display device that includes one or more buttons and/or a touchscreen for receiving input from a user. In some implementations, the input/output deviceincludes a keyboard and/or a pointing device. In some implementations, the input/output deviceis located within a cab of a refuse collection vehicle (e.g., within cabof vehicle). For example, the input/output devicecan be attached to or incorporated within a dashboard inside the cab of a refuse collection vehicle.

While certain embodiments have been described, other embodiments are possible.

115 136 115 136 115 136 115 115 115 115 102 115 For example, while the method of determining the number of remaining refuse containers to be collected has been described as including the steps of dividing the current SOC of the battery pack(s)by the rolling average battery usage, other methods of determining the remaining number of “picks” in a refuse collection route are possible. For example, in some embodiments, the method of determining the estimated remaining number of remaining refuse containers to be collected in a refuse collection route includes the onboard computing devicedetermining the charge of the battery pack(s)that is used per “pick” for every refuse container that is collected and then calculating an average. In some embodiments, once the average charge per “pick” is calculated, the onboard computing devicedetermines an estimated remaining number of remaining refuse containers to be collected by dividing the current SOC of the battery pack(s)by the average charge per “pick.” In some embodiments, the onboard computing devicedetermines the charge of the battery pack(s)by measuring the amount of energy of the battery pack(s)that is used per refuse container that is collected. In some embodiments, the amount of energy is measured in kilowatt-hours (kWh). In some embodiments, the amount of energy is calculated based on voltage measurements. In some implementations, the voltage measurements of each battery cell of the battery pack(s)are provided by the manufacturer of the batteries. In alternative embodiments, the voltage measurements of each battery cell of the battery pack(s)are measured by an operator or by a device onboard the vehicle. In some embodiments, this alternative method of determining the estimated remaining number of remaining refuse containers to be collected can advantageously exclude the contribution of additional vehicle components (e.g., the electrically-actuated refuse packing assembly or chassis) to the depletion of battery power and the rolling average battery usage. Furthermore, in some embodiments, this alternative method of determining the estimated remaining number of remaining refuse containers to be collected can advantageously remain unbiased to a battery pack or cell having a variable rate of discharge. For example, if a battery pack(s) discharges at a variable rate (e.g., faster or slower than average), the rolling average battery usage would be affected, whereas determining the average charge of the battery pack(s)that is used per “pick” for every refuse container that is collected would not be affected.

Although the following detailed description contains many specific details for purposes of illustration, it is understood that one of ordinary skill in the art will appreciate that many examples, variations and alterations to the following details are within the scope and spirit of the disclosure. Accordingly, the exemplary implementations described in the present disclosure and provided in the appended figures are set forth without any loss of generality, and without imposing limitations on the claimed implementations.

Although the present implementations have been described in detail, it should be understood that various changes, substitutions, and alterations can be made hereupon without departing from the principle and scope of the disclosure. Accordingly, the scope of the present disclosure should be determined by the following claims and their appropriate legal equivalents.

The singular forms “a,” “an,” and “the” include plural referents, unless the context clearly dictates otherwise.

As used in the present disclosure and in the appended claims, the term “state-of-charge” or “SOC” is defined as the ratio of the available battery pack(s) capacity or charge and the maximum possible capacity or charge that can be stored in the battery(s). In some examples, a fully charged battery pack(s) has an SOC of 100% while a fully discharged battery pack(s) has an SOC of 0%.

102 As used in the present disclosure, the term “pick(s)” is defined as the collection of a refuse container by the vehiclein a refuse collection route.

As used in the present disclosure and in the appended claims, the words “comprise,” “has,” and “include” and all grammatical variations thereof are each intended to have an open, non-limiting meaning that does not exclude additional elements or steps.

As used in the present disclosure, terms such as “first” and “second” are arbitrarily assigned and are merely intended to differentiate between two or more components of an apparatus. It is to be understood that the words “first” and “second” serve no other purpose and are not part of the name or description of the component, nor do they necessarily define a relative location or position of the component. Furthermore, it is to be understood that the mere use of the term “first” and “second” does not require that there be any “third” component, although that possibility is contemplated under the scope of the present disclosure.

Ranges may be expressed herein as from “about” one particular value and/or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. The use of the term “about” in this disclosure, when used to describe a numerical range or value, references a margin within +5% of the stated value or range. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it will be understood that the particular value forms another embodiment. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint.